The Geography of Economic Mobility in 19th Century Canada

Luiza Antonie, University of Guelph
Kris Inwood, University of Guelph
Chris Minns, London School of Economics and Political Science (LSE)
Fraser Summerfield, St. Francis Xavier University and RCEA

Geographically extended economies may exhibit considerable regional differences in economic performance leading to inter-regional differences in intergenerational economic mobility. In this paper we use linked Census data spanning 1871 to 1901 to provide the first portrait of local intergenerational mobility in Canada circa 1900. We construct multiple measures of both occupational and earnings mobility by 1871 Census sub-district, and document patterns of intergenerational mobility, their correlates at a local level, and the empirical relationships between these measures. We find the the highest rates of occupational mobility in Ontario and the Maritimes; the lowest are in Quebec and among Francophones. A more fine-grained analysis using 158 census divisions and GIS software allows us to illustrate and estimate the importance of local district characteristics. Agricultural districts, and locations with fewer interprovincial or international migrants had less occupational mobility as did districts with more Catholics and more French Canadians. More surprising is that a higher share of literate adults was associated with lower occupational mobility, though the share of children in education had a positive correlation with mobility. Analysis of relative income mobility shows some parallels, with rank-rank elasticities that were significantly higher (less mobility) in districts with less internal migrants or foreign migrants, and with higher literacy rates. Sons had higher predicted ranks of absolute mobility in areas that were more urbanized, where agriculture was as smaller share of economic activity, with more children in school, and more international and internal migrants. We find no notable correlation between the share Catholic or Francophone and absolute mobility. At first glance, locations with more occupational mobility typically experienced more absolute income mobility, although this relationship was mediated by local economic conditions. Finally, we match 1871 Census sub-districts to 1986 Census divisions in order to address change in mobility patterns over the 20th century

No extended abstract or paper available